UUID to HANA column type mapping

[pradu2]

Hi,

I'm having issues with the mapping of UUID to the native .

My changeset is something like this :

<changeSet id="mt-base-0001" author="initial">
 	<createTable tableName="some_table" >
 		<column name="id" type="UUID">
 			<constraints nullable="false" primaryKey="true" primaryKeyName="some_table_pk"/>
 		</column>
        </createTable>
<changeSet>

Database is HANA 2, on-prem.
The id column is created as VARCHAR(36).
As the ids are generated by Hibernate, it sometimes fails, with exception SAP DBTech JDBC: [274]: inserted value too large for column: Failed in "ID" column with the value .... .

As a workaround, I created a custom UUIDType with higher priority which maps UUID to VARBINARY(16).

Shouldn't this be also the mapping in this library, as the core or other dialects map it to either a UUID type (core, H2, Postgres)?

Note: I know that I could put there VARBINARY(16) as a type in the changeSet, but we're using H2 and Postgres in parallel and it would have invalidated the common changelog hashes.

Kind regards,

Radu

┆Issue is synchronized with this Jira Bug by Unito